Studies on stand dynamic growth model for larch in Jilin in China

The stand growth and yield dynamic models for Larch in Jilin Province were developed based on the forest growth theories with the forest continuous inventory data. The results indicated that the developed models had high precision, and they could be used for the updating data of inventory of planning and designing and optimal decision of forest management.

Machine learning-based real-time visible fatigue crack growth detection

Many large-scale and complex structural components are applied in the aeronautics and automobile industries.However,the repeated alternating or cyclic loads in service tend to cause unexpected fatigue fractures.Therefore,developing real-time and visible monitoring methods for fatigue crack initiation and propagation is critically important for structural safety.This paper proposes a machine learning-based fatigue crack growth detection method that combines computer vision and machine learning.In our model,computer vision is used for data creation,and the machine learning model is used for crack detection.Then computer vision is used for marking and analyzing the crack growth path and length.We apply seven models for the crack classification and find that the decision tree is the best model in this research.The experimental results prove the effectiveness of our method,and the crack length measurement accuracy achieved is 0.6 mm.Furthermore,the slight machine learning models help us realize real-time and visible fatigue crack detection.

Prediction of fatigue crack growth rate for smallsized CIET specimens based on low cycle fatigue properties

Based on experiments of low cycle fatigue for 5083-H112 aluminum alloy, two energybased predictive models have been introduced to predict the fatigue crack growth behaviors of traditional Compact Tension（CT） and small-sized C-shaped Inside Edge-notched Tension（CIET）specimens with different thicknesses and load ratios. Different values of the effective stress ratio U are employed in the theoretical fatigue crack growth models to correct the effect of crack closure.Results indicate that the two predictive models show different capacities of predicting the fatigue crack growth behaviors of CIET and CT specimens with different thicknesses and load ratios.The accuracy of predicted results of the two models is strongly affected by the method for determination of the effective stress ratio U. Finally, the energy-based Shi＆Cai model with crack closure correction by means of Newman＇s method is highly recommended in prediction of fatigue crack growth of CIET specimens via low cycle fatigue properties.

Investigating fatigue behavior of gear components with the acoustic emission technique

A novel method is presented to evaluate the complicated fatigue behavior of gears made of20Cr2Ni4 A.Fatigue tests are conducted in a high-frequency push-pull fatigue tester,and acoustic emission（AE）technique is used to acquire metal fatigue signals.After analyzing large number of AE frequency spectrum,we find that：the crack extension can be expressed as the energy of specific frequency band,which is abbreviated as F-energy.To further validate the fatigue behavior,some correlation analysis is applied between F-energy and some AE parameters.Experimental results show that there is significant correlation among the Fenergy,root mean square（RMS）,relative energy,and hits.The findings can be used to validate the effectiveness of the F-energy in predicting fatigue crack propagation and remaining life for parts in-service.F-energy,as a new AE parameter,is first put forward in the area of fatigue crack growth.

ERROR GROWTH IN NUMERICAL PREDICTION AND ATMOSPHERIC PREDICTABILITY

The article is to report some results of numerical experiments on the error growth and the atmospheric predictability Experiments with two-level global baroclinic primitive equation spectral model have main results as follows.The magnitude of initial errors directly affects the error growth,but its distribution form has little effect on the growth.The loss of predictability resulting from small-scale error is much greater than that from large-scale error.The small-scale error rapidly grows and is transferred to the large-scale error by interaction between different scale waves,which stimulates the growth of error for the whole system Orographic forcing restrains planetary-scale error(wavenumbers 0—3)but enhances the small-scale error (wavenumbers 8 or greater).Hence,orographic effects on the error growth closely depend on the characteris- tic scale of initial errors,and there may be a critical wavenumber between 4 and 7.The error growth is great- er in Northern Hemisphere than in Southern Hemisphere if initial errors are the same.In the end we give some discussions about model,initialization scheme,etc.,to improve model prediction.

User Profiling for CSDN:Keyphrase Extraction,User Tagging and User Growth Value Prediction

The Chinese Software Developer Network(CSDN)is one of the largest information technology communities and service platforms in China.This paper describes the user profiling for CSDN,an evaluation track of SMP Cup 2017.It contains three tasks:(1)user document keyphrase extraction,(2)user tagging and(3)user growth value prediction.In the first task,we treat keyphrase extraction as a classification problem and train a Gradient-Boosting-Decision-Tree model with comprehensive features.In the second task,to deal with class imbalance and capture the interdependency between classes,we propose a two-stage framework:(1)for each class,we train a binary classifier to model each class against all of the other classes independently;(2)we feed the output of the trained classifiers into a softmax classifier,tagging each user with multiple labels.In the third task,we propose a comprehensive architecture to predict user growth value.Our contributions in this paper are summarized as follows:(1)we extract various types of features to identify the key factors in user value growth;(2)we use the semi-supervised method and the stacking technique to extend labeled data sets and increase the generality of the trained model,resulting in an impressive performance in our experiments.In the competition,we achieved the first place out of 329 teams.

Reliability Growth Test Planning and Verification of Commercial Vehicles

Reliability and durability are two important technical indicators in automobile research and development.A research-and-design and testing organization can increase inherent quality attributes by adopting a systematic approach based on statistical tools and clearly defined processes.The process affects the design phase,validation through testing,and quality assurance in production.On the basis of reliability growth theory and the Duane model,this study established an estimation method for the definition of the target mileage and specific test cycles in reliability growth testing.A construction method for defin-ing test conditions was proposed that adopts the theory of the design of experiments.The simulation was conducted under a variety of typical test conditions including differing operation times,loads,and logistics modes to predict customer use and detect failures.Failure cases were then analyzed in detail.At the same time,a reliability growth prediction model was established on the basis of the initial test data and used for test process tracking and risk control.

Dust accumulated fungi in air-conditioning system:Findings based on field and laboratory experiments

This study analyzes the growth and reproduction of dust accumulated fungi(DAF)in an air-conditioning system based on field measurement and molecular biology,laboratory experiment and prediction modelling.The field measurement was conducted to collect dust in filter screen,surface cooler and air supply duct of two air handling units(AHUs).The results indicate that dust volume and fungal number in two AHUs generally met the hygienic specification of public buildings,but the cleansing did not fulfil requirements.High-throughput sequencing was conducted,revealing that the dominant fungal species were Alternaria_betae-kenyensis,Cladosporium_delicatulum,Aspergillus_sydowii,Verticillium_dahliae.Laboratory experiment was conducted to analyze the impact of several factors(e.g.growth time,temperature,relative humidity,duct material)and their combination on the DAF growth.The results indicate that fungal growth increased with time,peaking at 4 days or 5 days.Higher relative humidity or temperature was conducive to fungal growth.The orthogonal experiment revealed that the condition of“antibacterial composite,22±1℃and 45%-55%RH”had the strongest inhibiting impact on fungal growth.Logistic model,Gompertz model and square-root model were further developed to predict the fungal growth under different conditions.The results show that the Logistic model had high feasibility and accuracy,the Gompertz model was feasible with lower accuracy and the square-root model was feasible with high accuracy.Overall,this study facilitates the understanding of the DAF growth in air-conditioning ducts,which is important for real-time prediction and timely control of the fungal contamination.